Three themes relating to descriptions of chronic pain emerged from our analysis: diversity in conceptualising pain using a simple numerical score; personalising the meaning of pain by way of stories, similes and metaphors; and, contextualising pain in relation to its impact on everyday activities. As in our study, others have reported that the variation and unpredictability of pain makes it hard to rate numerically, and not just amongst the older population [25, 26]. There is a viewpoint that questions the meaningfulness and value of ascribing a number to complex experiences such as pain . Nevertheless, simple scales, including the numerical rating scale used in this study, are widely recommended and have provided much useful information [28, 29]. The findings from our study suggest that the simple numerical scale used here may not provide the scope for some older adults to adequately convey some of the features, variation and potential outcomes that they see as important . There are a number of pain assessment scales that are purported to be effective when assessing pain in older adults such as numerical rating scales or verbal descriptors (BPS/BGS Guidelines for the Assessment of Pain in Older Adults 2007 . Nonetheless, it is acknowledged that such scales only measure one aspect of pain which is intensity: it is essential to adopt a holistic, multi-dimensional approach to assessment to capture more of the complexity of pain; such as using a variety of terms synonymous with pain to screen older adults . Cicely Saunders introduced the concept of ‘total pain’ which places emphasis on the person and his or her experience of illness; recognising that their pain might arise from physical, psychological, spiritual or social distress, rather than simply referring to the body and its manifestations of disease . An individual’s concern, for example, about delays in diagnosis may impact upon their physical experience of pain. Efforts should therefore be made to provide scope for storied accounts that have the potential to provide contextual information that is more personally meaningful; inadequate exploration of pain is likely to be detrimental to comprehensive assessment of needs and thus poor management of symptoms and evaluation of treatment [32–35].
Our findings demonstrate the value of describing pain verbally in the context of the individual’s life [13, 14]. Older adults were inclined to construct a narrative around their pain, even in response to ‘closed’ questions, and used this as the basis for much of their interviews, to give their pain some context and meaning. This is a similar finding to that of Gooberman-Hill and colleagues , who used a different method, the ‘questerviews’ technique, to elicit older adults’ descriptions and experiences of joint pain in the context of standard self-assessment questionnaires (which included the Hip Disability and Osteoarthritis Outcome Score). They concluded that the assessment or measurement of pain should take into account the importance of pain experience, as well as severity, through use of patient narrative accounts. In our study, participants also related their pain to physical or social activity as a means of indicating its severity and impact; for example, their ability to walk, on social activities and/or having to adapt these activities. This suggests that ‘measuring’ the impact of pain against activities that are important to the older person may be another approach to gauging the effect of pain management; for example, by asking, following treatment, if they are able to walk any further with their dog, or about any increased social activity. Sometimes, the use of similes and metaphors provided a useful vehicle to overcome initial difficulties in describing pain and was used to indicate its severity.
Current UK policy states that practitioners should work in partnership with older adults in order that early intervention, assessment and management of chronic pain is founded on the older individual’s needs and wishes . This can only be achieved if we are willing to listen to older adults’ accounts of chronic pain. In this respect, chronic pain medicine may learn from palliative medicine, where, attending to the central message of individuals’ stories in terms of improving understanding of their total needs has a long tradition .
In practice, this requires allocation of time and a willingness to listen. Dillon et al  used a post-test only, randomised, double-blind experiment to test how the phrasing of health practitioners’ pain questions affected the pain information provided by community-living older adults with chronic osteoarthritis pain. They found that phrasing of the practitioners’ pain questions significantly impacted the amount of important pain information provided by older adults. Use of an open-ended request ‘tell me about your pain, aches, soreness or discomfort’ resulted in significantly more pain information than the closed-end question, ‘what would you rate your pain, aches, soreness or discomfort on a 0 to 10 scale, with 0 no pain, and 10 the worst pain possible?’ or the more social desirability-biased question, ‘how are you feeling?’ Social desirability refers to individuals responding in ways that are perceived to be the most socially acceptable rather than reflecting how they actually feel. Dillon and colleagues  concluded that practitioners should routinely ask open-ended pain questions when initiating pain assessments in order to provide a more tailored pain management approach for older adults. They also suggest that older adults should be helped to develop the most effective ways to communicate their pain to practitioners.
In terms of research approaches, seeking a more holistic assessment of pain can be more problematic in studies using quantitative designs; however, designs that include nested qualitative studies can provide this information. Methods that incorporate quantitative and qualitative approaches, such as the Day Reconstruction Method [39, 40] which uses structured diaries to collect data, may be worth further investigation and, as new technologies emerge to enable easier capture of rich objective and subjective data, such approaches may become more feasible in the future.
In common with much qualitative work, our findings should be treated as illustrative rather than statistically representative so there may be more issues in this field that have not emerged. Since we recruited through advertisement, it might be expected that people responded who wanted to talk about their pain and who did not have a straight forward answer to questions concerning their condition. The geographical location - North East Scotland - meant that possible influences and interplay of factors such as gender, ethnicity, socio-economic circumstances or sites of pain, could not be investigated in-depth and requires further study. The findings from the Chinese participants in particular, raised important questions about the assessment of people from ethnic backgrounds and whose first language is not English. However, a limitation is that the Chinese participants chose to be interviewed in a group, only once, and, at times, needed the help of an interpreter to clarify our questions and help them express their answers. This had an effect on the depth of data obtained; their answers tended to be short and probing to expand answers was difficult. It may therefore be seen as unsurprising that Chinese participants were able to ascribe a number to their pain, rather than detailed, verbal descriptions. In addition, it is possible that less confident participants may have deferred to the views of the more vocal members, although we did encourage everyone to speak in turn . This line of enquiry needs further investigation since health care professionals may need to use different approaches with different minority groups . The numerical scale was restricted to a simple numerical rating scale and so the findings do not necessarily relate to other measurement tools: that would be another useful extension of this study.